Primary Prevention of Obesity Through
Infancy-Based Interventions
Ian M. Paul, MD, MSc
Departments of Pediatrics & Public Health Sciences
Penn State College of Medicine & Children’s Hospital
Childhood Overweight Increasing Globally
Swinburn BA et al. Lancet, 2011
Why, What, How for Obesity Prevention
During Infancy
• Why is infancy a good time to intervene?
• What can be done to prevent obesity on an
individual level?
• How are we trying to demonstrate the efficacy of
interventions?
1963 - 2010
Most recent data from 2009-2010
2-5 year olds – 12.1 Obese, 26.7% overweight or obese
Ogden CL et al. JAMA, 2012.
Can We Interrupt the
Vicious Cycle of Obesity?
>25% of 2-5 year olds
are overweight
Overweight
Child /
Adolescent
X
Limited success
with interventions
Overweight /
Obese
Mother
↑Birth weight
or
Rapid infant growth
Opportunities for Prevention
40-50% of
women aged
20-39 years
Preventing Obesity – Why Infancy?
• Infancy represents an attractive period to intervene
–
–
–
–
period of great behavioral and metabolic plasticity
epigenetic changes in gene expression occur
numerous potential targets for preventive interventions
early life overweight and rapid infant weight gain are risk
factors for subsequent overweight and components of
metabolic syndrome
Birth Weight x Infant Weight Gain and
Risk of Obesity at age 7 years
Stettler N et al., Pediatrics, 2002. National Collaborative Perinatal Project (n=19,397)
What can be targeted?
Davison KK, Birch LL. Obesity Reviews, 2001
What can be targeted?
Environmental
Influences
Dietary Choices
for Infant
Maternal Biologic
Features
1. Pre-pregnancy BMI
2. Gestational Weight Gain
3. Insulin sensitivity and
Glucose Tolerance
1. Breastfeeding vs. Bottle Feeding
2. Timing of Introduction of Solids
3. Content of Diet
Infant Weight,
Weight Gain,
and Adiposity
Parent Behavior
1. Responsiveness to Infant Cues
(e.g. Hunger, Satiety, Fatigue,
Boredom, etc.)
2. Feeding Style
Nocturnal
3. Modeling Healthy Lifestyle
Feeding
Fetal & Neonatal
Biologic Features
1. Birth Weight
2. Genetic Predisposition
3. Hunger/Satiety Related
Hormones
E
Infant Behavior
1. Temperament and
Self-Regulation
2. Physical Activity and
Sedentary Behaviors
3. Sleep Duration
Traditional Child Feeding Practices
Evolved in Response to Food Scarcity
• Food:
– low palatability, low in energy & nutrients
– limited &/or unpredictable availability
– relatively expensive
• Feeding practices:
–
–
–
–
offer food to soothe crying
when available, provide large portions
offer palatable, liked foods if possible
pressure, force children to eat
• NIH/NIDDK-funded RCT with birth cohort intending
to breastfeed:
– SLeeping and Intake Methods Taught to Infants and
Mothers Early in Life (SLIMTIME) Study
• Two home nurse visits – 2-3 weeks, 4-6 months
after birth plus clinical research center visit at 1 year
Paul et al. Obesity, 2011
SLIMTIME Randomization Cells
Introduction of
Solids
Control
Soothe/Sleep
N = 42
2 interventions
N = 39
1 intervention
Control
N = 38
1 intervention
N = 41
0 interventions
SLIMTIME Intervention 1
• “Soothe/Sleep” instructions (delivered at 2-3 weeks)
– Infant Crying ≠ Hunger
– Parents must discriminate hunger vs. other distress
– Soothing strategy: 5 S’s (Swaddling, Side/Stomach,
Shushing, Swinging, Sucking)
– Day/night differences
• Key Measure
– Serial 96 hr diaries with 15 minute intervals recording
sleeping, awake/content, awake/fussy, feeding
Short sleep duration is a risk factor for
obesity in adults & children
Taveras EM et al., 2008
Infant night sleep duration increased for
BF mothers taught soothing techniques
Longest Sleep Bout
minutes
Intervention
Control
0
1
2
3 4 5 6
weeks after birth
7
8
Pinilla T and Birch LL. Pediatrics 1993; 91: 435-44
SLIMTIME Intervention 2
• “Introduction of Solids”
- delay introduction, hunger/satiety cues (2-3 wks)
- repeated exposure to vegetables (~4-6 mos)
• Key Outcome Measures
– age when solids introduced
– quantity of food consumed with repeated exposure
– acceptance of new food at age 1 year
• Primary outcome – Weight-for-length percentile at 1 yr
Repeated Exposure to Vegetables Overcomes
Infant Neophobia
1st Exposure
10th Exposure
• To overcome neophobia for lowenergy density foods, repeated
exposure is often required
Sullivan SA and Birch LL, 1994
Results - Demographics
• 110/160 (69%) completed the 1 year follow-up
• Infants completing study
– 51% female
– Mean birth weight – 3.33 kg (45th percentile for GA)
• Mothers completing study
– Mean age – 27.1 years
– 91% White, 90% married
– 65% completed college
Infant Feeding – Analysis Implications
• Study intended to study breastfeeding mother/baby pairs
• By 16 weeks only 50% still “predominantly breastfeeding”
– Defined as >80%* breast milk
*As done in second Infant Feeding Practice Study (IFPS II)
“Soothe/Sleep” Improves
Nocturnal Sleep for Breastfed Infants*
9
Mixed/Formula fed infants
(n=53; p=.40)
control
soothe/sleep
8
7
6
5
4
3
4
8
16
Hours of PM sleep (9 pm - 6 am)
Hours of PM sleep (9 pm - 6 am)
Breastfed infants
(n=55; p=.04)
9
control
soothe/sleep
8
7
6
5
4
3
Weeks
*Interaction p=.06
4
8
Weeks
16
“Soothe/Sleep” Reduces
Nocturnal Feeds for Breastfed Infants*
Mixed/Formula fed infants
(n=53; p=.94)
control
soothe/sleep
4
3
2
1
0
3
4
8
16
# of nightly feeds (9 pm - 6 am)
# of nightly feeds (9 pm - 6 am)
Breastfed infants
(n=55; p=.003)
4
control
soothe/sleep
3
2
1
0
3
Weeks
4
8
Weeks
*Interaction p=.04
16
“Sleep/Soothe” Reduces
Total Daily Feeds for Breastfed Infants*
Mixed/Formula fed infants
(n=53; p=.61)
control
soothe/sleep
11
10
9
8
7
6
5
4
3
4
8
16
Total # of feeds in a 24 hr period
Total # of feeds in a 24 hr period
Breastfed infants
(n=55; p=.008)
11
control
soothe/sleep
10
9
8
7
6
5
4
3
Weeks
*Interaction p=.05
4
8
Weeks
16
“Introduction of Solids” Intervention
Improves Timing of Introduction of Cereal
p=.06
“Introduction of Solids” Intervention –
Repeated Exposure at 4-6 months
Amount Consumed (g)
90
80
*
70
60
*
*
50
40
Day 1 Day 6
Week 1
*p<.05
Day 1 Day 6
Week 2
Day 1 Day 6
Week 3
Day 1 Day 6
Week 4
“Introduction of Solids” Improves
Acceptance of Unfamiliar Foods at 1 year
p=.05
Primary Outcome:
Weight-for-Length at age 1 year (N=110)
Secondary Outcomes: Alternatives to
feeding to soothe infant distress
• Diaries coded whether the infant was asleep,
fussing/crying, feeding, alert awake
• Markov time series models assessed differences in
intervention / control participant probability of infant
transitioning from crying to feeding vs. awake and alert
Anzman-Frasca SL et al. In progress, 2012
The Soothe/Sleep intervention increased infants’ probability
of transitioning from crying to awake/calm
Probability of transitioning from
fussing to awake at 16 wks
0.15
0.1
0.05
0
Control
Intervention
Soothe/Sleep study group
Infants’ probabilities of transitioning from fussing to feeding
positively predict their weight status at ~6 months
4
BMI-for-age z-scores
3
2
1
0
-1
-2
-3
-4
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Probability of transitioning from fussing to feeding
0.9
What else can be targeted?
Paul et al. Adv Pediatrics, 2009
Parents prefer higher percentiles
Laraway KA et al. Clin Pediatrics, 2010 and Sullivan SA et al. Clin Pediatrics, 2011
Obesity prevention is hard!
How can we put this together and
show that our approach is effective?
Moving Forward with a Conceptual
Framework: Responsive Parenting
• Coming from developmental literature, responsive
parenting involves prompt, emotionally supportive,
contingent, and developmentally appropriate
responses to infant cues
• Responsive parenting associated with:
– Secure attachment
– Language development
– Cognitive development
Responsive Feeding
• Prompt, contingent, developmentally appropriate
responses to infant/toddler hunger/satiety cues
• Shared feeding responsibility – “parents provide,
children decide”
• Fosters development of self-control in feeding and
self-regulation of eating
Discordant Feeding Responsiveness Æ
Overweight / Obesity
• Controlling, restrictive, or
coercive feeding
attenuates children’s
responsiveness to
hunger/satiety cues
– eating in absence of
hunger
– preferences for energy
dense foods
– increased obesity risk
DiSantis KI et al. Int J Obesity, 2011
The Intervention Nurses Start Infants
Growing on Healthy Trajectories
(INSIGHT) Study
INSIGHT Design
• Randomized, controlled trial with birth cohort (N=276):
– “Parenting” intervention vs. Child Safety Control
• Primary outcome – BMI at age 3 years
Participants – Inclusion / Exclusion
•
•
•
•
•
•
Singleton, term newborns ≥ 37 weeks gestion
Birth weight ≥ 2500 grams
Primiparous mothers ≥ 20 years old
English speaking
No major maternal/infant morbidities
Breastfeeding or formula feeding
Responsiveness in various domains
• Sleep & Drowsy
– establish good sleep hygiene while being responsive
without feeding as first sign of distress at night
– avoid feeding to sleep
• Fussy, but awake
– understand individual temperament differences
– don’t assume all fussiness is due to hunger, but rather
look for cues for hunger when fussy
– provide alternatives for feeding due to fussiness
Responsiveness in various domains
• Alert, calm, and feeding
– look for hunger/satiety cues
– “parents provide, children decide”
• Alert, calm, and playing
– minimize screen time
– limit sedentary/restrictive behaviors
– promote physical activity early in life
Additional elements
• Healthy nutrition choices
• Importance of parent modeling behaviors
• Education on normal growth and growth charts
Outcomes
•
•
•
•
•
•
Infant and maternal weight & BMI
Nutrition-related
Physical activity-related
Maternal mental health and parenting
Feeding style and feeding behaviors
Sleep
If this approach fails, our next attempt –
tape worms!